Latent class multinomial (or conditional) logit using R?

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
3 messages Options
Reply | Threaded
Open this post in threaded view
|

Latent class multinomial (or conditional) logit using R?

adanmartinez
Hi everyone?

Does anybody know how can I estimate a
Latent class multinomial (or conditional) logit using R?

I have tried flexmix, poLCA, and
they do not seem to support this model.
thanks in advance

adan
Reply | Threaded
Open this post in threaded view
|

Re: Latent class multinomial (or conditional) logit using R?

Christian Hennig
You may have a look at lcmixed/flexmixedruns in package fpc. This provides
flexmix methods that can be used to fit latent class models with locally
independent multinomials (no logits, though, and I'm not sure that for
categorical data only this is much different from what poLCA offers).

Best regards,
Christian

On Fri, 23 Dec 2011, adanmartinez wrote:

> Hi everyone?
>
> Does anybody know how can I estimate a
> Latent class multinomial (or conditional) logit using R?
>
> I have tried flexmix, poLCA, and
> they do not seem to support this model.
> thanks in advance
>
> adan
>
> --
> View this message in context: http://r.789695.n4.nabble.com/Latent-class-multinomial-or-conditional-logit-using-R-tp4230083p4230083.html
> Sent from the R help mailing list archive at Nabble.com.
>
> ______________________________________________
> [hidden email] mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>

*** --- ***
Christian Hennig
University College London, Department of Statistical Science
Gower St., London WC1E 6BT, phone +44 207 679 1698
[hidden email], www.homepages.ucl.ac.uk/~ucakche

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Reply | Threaded
Open this post in threaded view
|

Re: Latent class multinomial (or conditional) logit using R?

adanmartinez
Hi Christian,
Thanks a lot for the suggestion.
As you expected, lcmixed does not provide more functionality than poLCA for the case of
only categorical data...
Do you know whether there is an explanation (some technical difficulty I
may be missing, I am Economist so it is very likely) of why I can fit a Poisson, a logit latent class
model but not a conditional logit?

In fact, I started using R when I realize that my (very rudimentary) EM script written in Matlab
had tons of problems to converge.
Now I start wondering if there is some specific situation I should be aware of.

Thanks a lot again